Heat Dissipation of High Power Electronic Devices

High-power electronics are at the heart of applications such as electric vehicles, wind turbines, high-speed rail, and power grids. At present, high-power electronic devices are developing toward high power level and high integration. Therefore, the problem of heat dissipation is inevitably concerned. The heat generated by high-power semiconductor devices will cause the temperature of the chip to rise. Without proper heat dissipation measures, the operating temperature of the chip will exceed the maximum allowable temperature. This will lead to deterioration or even damage of device performance. Studies have shown that for every 10°C increase in the temperature of a semiconductor chip, the reliability of the chip will be reduced by half. The higher the operating temperature of the device, the shorter the lifetime of the device. Reducing the temperature of a device is an effective way to extend its life cycle. At this stage, there have been many researches on thermal design and heat dissipation optimization of various electronic devices and equipment.

The impact of temperature on the life of electronic devices is mainly reflected in two aspects. One is thermal failure of the chip, and the other is stress damage. The safe working temperature of common silicon chips is generally - 40 ~ 50 ℃, and the device can work normally within the safe working temperature range. When the junction temperature exceeds the safe operating temperature, it will cause thermal failure of the chip. The maximum allowable junction temperature for silicon chips is generally 175°C. On the other hand, due to the difference in the expansion coefficient of each material in the device, an excessively high junction temperature will cause an increase in thermal stress in the chip. This in turn causes mechanical damage such as bending of the solder in the chip and falling off of the bonding wire. Some scholars have pointed out in their research that for flip chips on lead frames, due to the large difference in thermal expansion coefficient between the copper lead frame and the silicon chip in the package, the thermal stress under the action of thermal load will cause damage to the surface structure of the chip connected to the solder joints. . Some scholars also pointed out in the article that the residual stress generated by the traditional reflow soldering interconnection technology used in semiconductor device packaging will be further aggravated at high temperature. Eventually lead to brittle fracture of the chip and the solder layer of the substrate. In addition, an excessively high junction temperature can also cause thermal breakdown of the chip, or even thermal melting of the chip. These failures are non-recoverable failures, so the damage of high temperature to the device is fatal.

The parameters of the electronic device itself are very sensitive to temperature changes. Its on-state resistance, forward voltage drop, threshold voltage, conduction current and other parameters will change with the change of temperature. For example, the on-state resistance of a power MOSFET increases approximately linearly with increasing junction temperature. Therefore, the homomorphic loss of the device will also increase, causing more heat to be generated by the device. This further increases the junction temperature, creating a vicious cycle. For IGBT, relevant studies have shown that the turn-off delay time will increase with the increase of the operating junction temperature of the device. Reasonable use of thermal parameters can be used as a characterization parameter of device junction temperature. The loss of control of thermal parameters can cause serious damage to the device, and the damage caused by thermal parameters tends to deteriorate further with the increase of temperature.

As the core components of power electronic equipment, power electronic devices will inevitably generate various losses during operation, including conduction loss and switching loss. If the heat generated by the device is not dissipated to the surrounding environment in time, the high operating temperature will seriously affect the normal operation of the device and the reliable operation of the device. With the advancement of power electronics technology, equipment is developing in the direction of miniaturization and compactness. This makes the characteristics of heat concentration and small heat dissipation area of power electronic devices increasingly prominent, resulting in the continuous increase of surface heat flux density of devices. In high-power applications, it is necessary to install additional heat sinks to achieve reliable operation of the equipment. In addition, with the application of new materials such as SiC in power electronic devices, due to the reduction of chip size, the local heat flux density is higher, and the heat dissipation requirements are higher.

With the rapid development of power electronic devices, the application scenarios are constantly expanding and complex. Equipment usually faces various external environments such as high temperature, high humidity, high salt, vibration and even vacuum. This subjects the device and the components inside it to various tests. At the same time, higher requirements are placed on the heat dissipation system of the equipment. Therefore, it is necessary to consider the influence of different environmental parameters on the thermal design of the device. Some scholars have analyzed the development needs of power electronics technology at extreme temperatures, pointing out that extreme high temperature and low temperature environments in aerospace and other fields are inevitable for power electronic devices. Therefore, the performance research of devices in extreme environments is very important. Some scholars have considered the characteristics of high temperature and high humidity in the coal mine environment, and analyzed the temperature rise characteristics of the mine motor power converter in a humid environment. Through simulation and experimental research, it is found that under the same ambient temperature, the maximum temperature rise of the power converter decreases with the increase of the relative humidity of the environment. This has a preliminary understanding of the law of heating in humid environments. Due to the complexity of the application environment, in the design of power electronic equipment, not only the influence of the environment on internal components must be considered, but also the particularity of the thermal design of the equipment. According to different environmental characteristics, optimize the heat dissipation method. In addition, it is also necessary to consider the influence of the environment on the heat dissipation system to improve the heat dissipation efficiency and reliability of the heat dissipation system.

The thermal design of power electronics involves more than just the field of heat transfer. As shown in Figure 1, in the heat transfer path analysis of typical power electronic devices using the thermoelectric simulation method, in order to achieve a good heat dissipation effect and take into account the reliability, light weight and miniaturization requirements of the equipment, it is necessary to comprehensively consider the temperature field and the stress field Coupling with the flow field. It can be seen from the above analysis that the thermal design of power electronic equipment is a research involving many disciplines such as mechanics, electronics, heat transfer and fluid mechanics. Therefore, it is necessary to consider the mechanical-electrical-thermal integration design of power electronic devices, and focus on the coupling of electric-thermal-mechanical multi-physics fields of power electronic devices.

The heat transfer process of power electronic devices includes three ways: heat conduction, heat convection and heat radiation. Heat conduction from the chip to the heat sink and heat convection from the heat sink to the surrounding environment are the main heat transfer methods. The heat dissipation design of power electronic equipment mainly starts from these two aspects. Common heat dissipation methods can be divided into active heat dissipation, passive heat dissipation, and thermoelectric cooling according to the way the heat is taken away from the radiator. Passive heat dissipation mainly includes common natural convection, indirect contact gas-liquid, solid-liquid phase change cooling, and direct contact immersion liquid cooling and phase change cooling. Active cooling mainly includes common forced air cooling and forced liquid cooling. In order to give full play to the heat dissipation capacity of the existing heat dissipation methods, the heat dissipation technology of power electronic equipment is constantly optimizing and improving the existing heat dissipation methods while developing new heat dissipation technologies. Figure 2 is a schematic diagram of the range of heat flux corresponding to common heat dissipation methods.

Natural convection heat dissipation technology uses air as heat transfer medium. It uses the buoyancy generated by the thermal expansion and cold contraction of air itself to make the air flow around the radiator fin to realize the exchange between hot air and cold air. Compared with other heat dissipation methods, natural convection heat dissipation requires no additional energy, simple structure, reliable operation and basically no maintenance. It is widely used in the situation of low heat flux. Because of the simple heat dissipation structure, the research on natural convection heat dissipation mainly focuses on the optimization of heat dissipation structure and installation direction. In recent years, there are many researches on heat dissipation supported by the theory of field synergy.

In contrast to natural convection heat dissipation, the movement of forced air-cooled cooled air is powered by a fan. Because the speed of the air is greatly increased, its heat dissipation ability is stronger. Its heat flux is obviously higher than natural convection heat dissipation, about 5 ~ 10 times of natural air cooling. The design of forced air cooling structure mainly includes the design of heat sink structure parameters, the selection of cooling fan and the design of fluid air duct. These three aspects of the design to make the heat dissipation area, air flow and air pressure drop balance, in order to make the forced air cooling heat dissipation play the best effect. Because the heat dissipation effect of forced air cooling is obviously better than that of natural air cooling, although the heat dissipation effect is not as good as that of forced liquid cooling, its complexity, volume, weight and late maintenance are obviously better than that of liquid cooling. It can be widely used and rapidly developed in the thermal design of high power electronic devices.

Figure 3 shows a typical structure of forced liquid cooling. The heat generated by the heat source in the heat dissipation structure is transferred to the cooling liquid through the device packaging and liquid cooling plate in the way of heat conduction. The heated liquid is transported to the heat ex-changer section under the action of the pump and the heat is dissipated to the surrounding environment through the heat ex-changer. Forced liquid cooling, in contrast to forced air cooling, transfers heat from the heat source to the heat ex-changer part by cooling the liquid. Direct contact with the heat source is liquid and liquid thermal conductivity is significantly higher than air, therefore, its heat dissipation effect is significantly better than forced air cooling heat dissipation, about 6 ~ 10 times of air cooling. In liquid cooling heat dissipation, the use of media with better thermal conductivity can significantly improve the heat dissipation effect. Some scholars put forward the application of liquid metal as cooling medium in the thermal development of the cooling system of power electronic devices, and verified the possibility of liquid metal applied in the liquid cooling heat dissipation of high-power power electronic devices through simulation and experiment. Due to the existence of liquid in the system, it is necessary to consider the replacement of liquid and prevent the damage of liquid leakage to the device. Forced liquid cooling has high requirements on liquid reliability and piping system. Due to the complex system structure and many components, the volume and weight of the system are obviously greater than the heat dissipation of air cooling. Therefore, the application environment of forced liquid cooling has certain restrictions.

Thermoelectric cooling takes advantage of the Partier effect of semiconductor materials, in which an electric current flows through the interface of two different materials, absorbing or releasing heat from the outside world. In recent years, with the development of semiconductor material manufacturing technology, thermoelectric cooling has developed rapidly. Figure 4 shows a typical structure of thermoelectric cooling. Although the cooling end of thermoelectric cooling can significantly reduce the temperature of the heat source, its total heat dissipation capacity is limited to that of the hot end. The heat dissipation effect of the whole system is closely related to the heat dissipation mode of the hot end. Due to the heat dissipation measures still need to be taken at the hot end of thermoelectric cooling, the overall heat dissipation system is complicated and cumbersome, which restricts its application.

Heat pipe heat dissipation is a liquid phase change heat transfer principle. The saturated liquid inside the heat pipe absorbs heat from the high temperature side and vaporizes. Saturated steam flows to the low temperature side to exotherm and condense into liquid, and returns to the high temperature side under the action of gravity or capillary force to continue to participate in the cycle of absorption and exotherm. Figure 5 shows the typical structure of gravity heat pipe. Although heat pipe heat dissipation is passive heat dissipation, it has excellent thermal conductivity which is incomparable to other metals. In recent years, various forms of heat pipe heat dissipation technology have developed rapidly.

There are two main definitions of micro-channels. A channel with a hydraulic diameter of 0.01 ~ 0.2 mm can be called a micro-channel. The other is defined by the ratio of buoyancy to surface tension. Regardless of the definition, micro-channel heat dissipation technology has attracted increasing attention from researchers due to its outstanding advantages such as small size, small heat transfer temperature difference and high heat transfer efficiency per unit area. In recent years, with the continuous improvement of micro-channel theory and the rapid development of processing technology, this technology has become a research hot topic of scholars. The research on micro-channel heat dissipation technology mainly focuses on channel size optimization and flow and heat transfer characteristics of channel medium.

With the deepening of research and development, new materials have been applied in various structural levels. Application of SiC band gap semiconductor materials in switching devices. Application of new phase change medium with high reliability in phase change heat dissipation. In thermal interface materials, various components of liquid metal applications.

In solving the problem of heat dissipation of high power electronic devices, the theory of thermodynamics should be based first. Starting with the fundamental laws of thermodynamics; Attach importance to the research and development of new materials and production. Whether it is heat dissipation material or thermal interface material, the new material has incomparable advantages. Develop new materials with superior thermal properties and reduce the cost of production and application, so that they can be widely used. This can give full play to the potential of heat dissipation technology and improve the heat dissipation effect. The research of new heat dissipation technology should also continue to be in-depth. In the development process of the existing heat dissipation technology from passive to active, from natural convection to forced air cooling to forced liquid cooling, and from single-phase heat dissipation to multi-phase heat dissipation, the heat flux has increased greatly. Although the new heat dissipation mode will inevitably be accompanied by the change of the overall structure, the increase of heat flux is significant, which is of great significance to improve the overall heat dissipation effect of the equipment.